Audio signals typically have a large dynamic range. For example, audio signals often have a peak-to-average magnitude ratio of 8-to-1 or even higher, which means that the average value of an audio signal is much smaller than possible peak values. However, the peaks in an audio signal are relatively infrequent and at most times an audio signal has a magnitude close to a relatively low average magnitude.
A power amplifier for amplifying an audio signal must be capable of producing an output signal with corresponding high peaks while most often producing an output audio signal with a comparatively low magnitude. Typically for a power amplifier, the DC bias current required to maintain the necessary operating range within the active amplifying devices to produce peak output power is significantly higher than the DC bias current required to amplify the audio signal at or close to the relatively low average level. Thus, given that peaks in an audio signal are infrequent, the bias current will be higher than required most of the time which in turn leads to low efficiency if peak power is always made available, such as in a typical class AB amplifier. In contrast it would be preferable if the available output power could be adjusted to track the requirements of the input signal. Those skilled in the art will appreciate that headroom is defined as the difference between the supplied power and that of the output audio signal.